InferenceObjects.jl is a Julia implementation of the InferenceData schema for storing results of Bayesian inference. Its purpose is to serve the following three goals:
- Usefulness in the analysis of Bayesian inference results.
- Reproducibility of Bayesian inference analysis.
- Interoperability between different inference backends and programming languages.
The implementation consists primarily of the
Dataset structures. InferenceObjects also provides the function
convert_to_inference_data, which may be overloaded by inference packages to define how various inference outputs can be converted to an
For examples of how
InferenceData can be used, see the ArviZ.jl documentation.